Nonparametric Tests of Conditional Treatment Effects

We develop a general class of nonparametric tests for treatment effects conditional on covariates. We consider a wide spectrum of null and alternative hypotheses regarding conditional treatment effects, including (i) the null hypothesis of the conditional stochastic dominance between treatment and control groups; (ii) the null hypothesis that the conditional average treatment effect is positive for each value of covariates; and (iii) the null hypothesis of no distributional (or average) treatment effect conditional on covariates against a one-sided (or two-sided) alternative hypothesis. The test statistics are based on L_{1}-type functionals of uniformly consistent nonparametric kernel estimators of conditional expectations that characterize the null hypotheses. Using the Poissionization technique of Gine et al. (2003), we show that suitably studentized versions of our test statistics are asymptotically standard normal under the null hypotheses and also show that the proposed nonparametric tests are consistent against general fixed alternatives. Furthermore, it turns out that our tests have non-negligible powers against some local alternatives that are n^{-1/2} different from the null hypotheses, where n is the sample size. We provide a more powerful test for the case when the null hypothesis may be binding only on a strict subset of the support and also consider an extension to testing for quantile treatment effects. We illustrate the usefulness of our tests by applying them to data from a randomized, job training program (LaLonde (1986)) and by carrying out Monte Carlo experiments based on this dataset.

[1]  Kyungchul Song.  Testing Conditional Independence via Rosenblatt Transforms , 2007, 0911.3787.

[2]  G. Harrison,et al.  Field experiments , 1924, The Journal of Agricultural Science.

[3]  Harshinder Singh,et al.  Testing for Second-Order Stochastic Dominance of Two Distributions , 1994, Econometric Theory.

[4]  I. Gijbels,et al.  Bandwidth Selection in Nonparametric Kernel Testing , 2008 .

[5]  Stephen G. Donald,et al.  Consistent Tests for Stochastic Dominance , 2003 .

[6]  Petra E. Todd,et al.  Matching As An Econometric Evaluation Estimator , 1998 .

[7]  Probal Chaudhuri,et al.  Nonparametric Estimates of Regression Quantiles and Their Local Bahadur Representation , 1991 .

[8]  Esther Duflo,et al.  WOMEN AS POLICY MAKERS: EVIDENCE FROM A RANDOMIZED POLICY EXPERIMENT IN INDIA , 2004 .

[9]  Monica Costa Dias,et al.  Alternative approaches to evaluation in empirical microeconomics , 2002, The Journal of Human Resources.

[10]  J. Hahn.  On the Role of the Propensity Score in Efficient Semiparametric Estimation of Average Treatment Effects , 1998 .

[11]  Rajeev Dehejia,et al.  Propensity Score-Matching Methods for Nonexperimental Causal Studies , 2002, Review of Economics and Statistics.

[12]  John E. Kolassa,et al.  Asymptotic Normality , 2011, International Encyclopedia of Statistical Science.

[13]  V. Chernozhukov,et al.  An IV Model of Quantile Treatment Effects , 2002 .

[14]  James J. Heckman,et al.  Econometric Evaluation of Social Programs, Part I: Causal Models, Structural Models and Econometric Policy Evaluation , 2007 .

[15]  E. Mammen,et al.  Identification and Estimation of Local Average Derivatives in Non-Separable Models Without Monotonicity , 2009 .

[16]  Richard K. Crump,et al.  Nonparametric Tests for Treatment Effect Heterogeneity , 2006, The Review of Economics and Statistics.

[17]  Xiaohong Chen,et al.  Semiparametric efficiency in GMM models with auxiliary data , 2007, 0705.0069.

[18]  J. Angrist,et al.  Instrumental Variables Estimates of the Effect of Subsidized Training on the Quantiles of Trainee Earnings , 1999 .

[19]  Edward Miguel,et al.  Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities, Guide to Replication of Miguel and Kremer (2004) , 2014 .

[20]  D. McFadden.  Testing for Stochastic Dominance , 1989 .

[21]  J J Heckman,et al.  Local instrumental variables and latent variable models for identifying and bounding treatment effects. , 1999, Proceedings of the National Academy of Sciences of the United States of America.

[22]  Sokbae Lee,et al.  Intersection bounds: estimation and inference , 2009, 0907.3503.

[23]  Subhashis Ghosal,et al.  TESTING MONOTONICITY OF REGRESSION By , 1998 .

[24]  O. Linton,et al.  Consistent Testing for Stochastic Dominance Under General Sampling Schemes , 2003 .

[25]  David Card,et al.  Estimating the Effects of a Time Limited Earnings Subsidy for Welfare Leavers , 2004 .

[26]  E. Giné,et al.  The L 1 – Norm Density Estimator Process , 2001 .

[27]  Rabindra N Bhattacharya.  On Errors of Normal Approximation , 1975 .

[28]  Esther Duflo,et al.  Women as Policy Makers: Evidence from a Randomized Experiment , 2010 .

[29]  Jeffrey A. Smith,et al.  Does Matching Overcome Lalonde's Critique of Nonexperimental Estimators? , 2000 .

[30]  P. Lavergne,et al.  An equality test across nonparametric regressions , 2001 .

[31]  Alberto Abadie.  Bootstrap Tests for Distributional Treatment Effects in Instrumental Variable Models , 2002 .

[32]  Oliver Linton,et al.  An Improved Bootstrap Test of Stochastic Dominance , 2009 .

[33]  Gordon Anderson,et al.  Nonparametric Tests of Stochastic Dominance in Income Distributions , 1996 .

[34]  J. Angrist,et al.  Identification and Estimation of Local Average Treatment Effects , 1995 .

[35]  Piotr Kokoszka,et al.  Testing for stochastic dominance using the weighted McFadden-type statistic , 2006 .

[36]  H. White,et al.  A NONPARAMETRIC HELLINGER METRIC TEST FOR CONDITIONAL INDEPENDENCE , 2008, Econometric Theory.

[37]  Alternative approaches to evaluation in empirical microeconomics , 2002 .

[38]  Kjell A. Doksum,et al.  On average derivative quantile regression , 1997 .

[39]  James J. Heckman,et al.  Econometric Evaluation of Social Programs, Part II: Using the Marginal Treatment Effect to Organize Alternative Econometric Estimators to Evaluate Social Programs, and to Forecast their Effects in New Environments , 2007 .

[40]  J. Heckman,et al.  Econometric Evaluation of Social Programs, Part III: Distributional Treatment Effects, Dynamic Treatment Effects, Dynamic Discrete Choice, and General Equilibrium Policy Evaluation , 2007 .

[41]  R. Lalonde.  Evaluating the Econometric Evaluations of Training Programs with Experimental Data , 1984 .

[42]  Quang Vuong,et al.  Nonparametric Identification of Risk Aversion in First‐Price Auctions Under Exclusion Restrictions , 2009 .

[43]  Myoung‐jae Lee.  Non‐parametric tests for distributional treatment effect for randomly censored responses , 2009 .

[44]  G. Imbens,et al.  Efficient Estimation of Average Treatment Effects Using the Estimated Propensity Score , 2000 .

[45]  V. Chernozhukov,et al.  Inference on Counterfactual Distributions , 2009, 0904.0951.

[46]  Wenceslao González Manteiga,et al.  Significance testing in nonparametric regression based on the bootstrap , 2001 .

[47]  D. Mason,et al.  Asymptotic normality of plug-in level set estimates. , 2009, 0908.1045.

[48]  Jianqing Fan,et al.  Robust Non-parametric Function Estimation , 1994 .

[49]  V. V. Shergin.  Central limit theorem for finitely-dependent random variables , 1993 .

[50]  R. Davidson,et al.  Statistical Inference for the Measurement of the Incidence of Taxes and Transfers , 1997 .

[51]  Jianqing Fan,et al.  Local polynomial modelling and its applications , 1994 .

[52]  C. Manski.  Statistical treatment rules for heterogeneous populations , 2003 .

[53]  R. Davidson,et al.  Statistical Inference for Stochastic Dominance and for the Measurement of Poverty and Inequality , 1998 .

[54]  O. Linton,et al.  Nonparametric Estimation of a Polarization Measure , 2009 .

[55]  Sergio Firpo.  Efficient Semiparametric Estimation of Quantile Treatment Effects , 2004 .

[56]  G. Imbens,et al.  Large Sample Properties of Matching Estimators for Average Treatment Effects , 2004 .

[57]  A. Cuevas,et al.  A plug-in approach to support estimation , 1997 .